Nguyen Minh Hoai, De la Torre Fernando
Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, USA.
Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2008 Jun 23;2008. doi: 10.1109/CVPR.2008.4587524.
Parameterized Appearance Models (PAMs) (e.g. Eigentracking, Active Appearance Models, Morphable Models) are commonly used to model the appearance and shape variation of objects in images. While PAMs have numerous advantages relative to alternate approaches, they have at least two drawbacks. First, they are especially prone to local minima in the fitting process. Second, often few if any of the local minima of the cost function correspond to acceptable solutions. To solve these problems, this paper proposes a method to learn a cost function by explicitly optimizing that the local minima occur at and only at the places corresponding to the correct fitting parameters. To the best of our knowledge, this is the first paper to address the problem of learning a cost function to explicitly model local properties of the error surface to fit PAMs. Synthetic and real examples show improvement in alignment performance in comparison with traditional approaches.
参数化外观模型（PAMs）（例如特征跟踪、主动外观模型、可变形模型）通常用于对图像中物体的外观和形状变化进行建模。虽然相对于其他方法，PAMs有许多优点，但它们至少有两个缺点。首先，它们在拟合过程中特别容易陷入局部最小值。其次，成本函数的局部最小值通常很少（如果有的话）对应于可接受的解决方案。为了解决这些问题，本文提出了一种方法，通过明确优化使得局部最小值仅出现在与正确拟合参数对应的位置来学习成本函数。据我们所知，这是第一篇解决学习成本函数以明确建模误差表面的局部属性以拟合PAMs这一问题的论文。合成和真实示例表明，与传统方法相比，对齐性能有所提高。
